Indication method for statistically occurring events The invention relates to a method for indicating the frequency of statistically occurring events and devices for using the method. In the explanations below, the detection of radioactive radiation is assumed as an example of statistically occurring events. However, it is pointed out that the invention is not restricted to the specific kind of statistically occurring events, but rather may also serve, by way of example, for indicating the current traffic density in road traffic or the stream of visitors to facilities.
An array of methods are known which are supposed to make it possible to represent the frequency of statistically occurring events in a meaningful manner.
Thus, the intensity of radioactive radiation is indicated by measuring devices with an analog direct-reading instrument by each pulse of the sensor being assigned a specific electric charge and the current resulting from the rate of charges which corresponds to the sensor pulse rate being conducted through the analog direct-reading instrument. If this direct-reading instrument is equipped with a sufficiently long mechanical time constant, then the resultant low-pass filtering produces an indication corresponding to the pulse rate.
The intensity of radioactive radiation is indicated by measuring devices with numerical indication by the number of pulses of the sensor being counted over a specific duration, for example one minute, and the number being indicated either continually or finally. The measurement time is often selected such that the indicated number of pulses corresponds to the rate with a unit selection.
Other devices do not have a time reference, but rather count the pulses continually from the beginning of a measurement to the end of a measurement, in order, for example, to be able to determine the total radiation load.
The published patent application 30 25 489 describes a device for measuring ionizing radiation which measures the time duration between the detected radiation pulses. Owing to the statistical behavior of ionizing radiation, the time interval between a plurality of radiation pulses is measured and the corresponding mean value is formed. The interval between two determinations of the mean value corresponds to the time interval between the first and last radiation pulse included in the averaging. Thus, a new measured value can be determined in a short time only at high rates.
U.S. Pat. No. 4,090,082 describes how, by taking account of the dead time of a Geiger-Mxc3xcller counter tube, it is possible to improve the linearity of Geiger-Mxc3xcller counters toward very high pulse rates and how the counting of pulses and its indication can expediently be effected. The dead time consideration is negligible or at least of secondary importance for Geiger-Mxc3xcller counter tubes provided for an application outside highly loaded radiation areas. Thus, the dead time in the case of the known counter tube LND712 is 90 microseconds; the typical pulse rate in the case of environmental radioactivity is in the region of less than one pulse per second. If, by way of example, motor vehicles traveling past are used as events to be recorded, then the dead time consideration acquires considerable importance owing to the relatively high event rate in that case for many sensors.
U.S. Pat. No. 4,837,705 describes a method for calculating the rate of change of the frequency of statistical events. This explains how difficult it is to ascertain how rapidly the neutron rate in a nuclear reactor is increasing or decreasing, because in order to determine it accurately, as is necessary for the purpose of controlling and monitoring nuclear power stations, a long measurement duration is necessary and also tenable in the case of a low frequency of statistical events, while accurate determination is possible rapidly and is also necessary in the case of a high frequency.
The methods known to date are affected by many problems which limit their use or even make them dangerous to use. Thus, although analog measuring devices are still suitable for fast indication, they are not suitable for accurate evaluation or for long-term integration of the measured radiation. The numerical measuring devices based on counters are not suitable for fast indication of radiation that has increased to a great extent. For this reason, they are often combined with acoustic indication, which is intended to inform the observer about the present intensity of the radiation.
For the purpose of temporally resolved logging, a number of measuring devices are provided with an analog output which supplies a signal that corresponds to the current pulse rate and can be recorded. One disadvantage of such logging is that the handling of the measurements is made more difficult and its evaluation becomes costly and complicated.
In the case of the methods used in nuclear power stations, the technical outlay for determining indication values is of only secondary importance. Thus, U.S. Pat. No. 4,837,705 specifies that, by way of example, the pulses are in each case counted for a tenth of a second and 16,640 such counts are stored and used for determining measured values.
A further problem is that although the measuring devices indicate the current or the averaged rate of the radiation, they do not indicate the permissible residence duration linked to the indicated rate. The sensitivity of Geiger-Mxc3xcller counter tubes used as sensor is usually specified in pulse rate per dose, for example in 1.8 pulses per second for each microsievert per hour. This corresponds to 6480 pulses per microsievert. The permissible annual dose for personnel professionally exposed to radiation is for example 50 millisievert in Germany, from which the sensor pulse number corresponding to this annual dose can be determined in a customary manner, 324,000,000 pulses in the example. The permissible residence duration results, as is generally known, from this pulse number referring to a year and divided by the currently measured pulse rate. Thus, in the example, a pulse rate of about 20 pulses per second produces a permissible residence duration of half a vear.
The object of the invention is to avoid the disadvantages of the known methods and devices and to make it possible that firstly, on the one hand, indication values which are as accurate as possible are displayed as fast as possible and, secondly, the rate and the permissible residence duration are indicated.
According to the invention, the above object is achieved by means of a method according to claim 1. In order to indicate the rate of the pulses output by the sensor, the rate corresponding to the rate of recorded events, for each individual pulse the current time period is in each case defined as a temporal feature and either the latter is stored and the indication values and their representation are calculated from the stored features or, for each time period, the pulses assigned thereto are counted and these numbers are stored and the indication values and their representation are calculated therefrom. For simultaneous graphical representation of the pulse rate and the permissible residence duration, a nonlinear, for example a logarithmic, graphical indication of the pulse rate in a bar with dual scaling, in opposite directions, for the pulse rate and the permissible residence duration is performed.
In one refinement of the invention, the point in time at which the pulse occurs, for example with a resolution of one second, is determined as the temporal feature of said pulse and is stored in a circulating memory having a specific number of memory locations. At the respective point in time of a desired indication, the difference between the current time and the oldest time stored in the circulating memory is formed and the number of memory locations is divided by this and indicated as the rate, if appropriate scaled with a factor. Furthermore, the number of pulses is counted in a counter and stored for logging purposes at intervals with the current temporal feature.
The position of the oldest time in the circulating memory corresponds to that at which the temporal feature of the next pulse is to be entered. The essential advantage of this refinement is that even given a low rate, an indication with indication values having relatively little variation is achieved and, if the radiation increases, the indication immediately increases and reaches its final steady-state value with a shorter and shorter averaging time.
When a device according to the invention is switched on, the indicated value is incorrect until the circulating buffer has been filled for the first time.
In a first further refinement of the invention, therefore, until the circulating buffer has been completely filled for the first time, the switch-on point in time is used as the oldest time and the current number of used memory locations is used as the number of memory locations. This means that the indication initially varies more greatly owing to the initially smaller number of pulses which contributes to the averaging.
In a second further refinement of the invention, when the measuring device is switched on, in accordance with the expected rate, pre-allocation of the memory locations is carried out with previous temporal features corresponding to the rate. As a result, the variation of the indication corresponds to steady-state behavior even directly after the switch-on.
In one refinement of the invention, a time period whose formation is based on a clock cycle, for example with a clock period of one second, is selected as the temporal feature of the pulse, and the numbers of pulses occurring per second are stored in a circulating memory having a specific number of memory locations. At the respective point in time of a desired indication, the sum of the numbers stored in the memory locations is divided by the number of memory locations and indicated as the rate, if appropriate scaled with a factor. If the number of memory locations is selected skillfully, for example 64 if binary numbers are used or for example 100 if decimal numbers are used, point shifting is sufficient so that explicit division can be dispensed with. The essential advantage of this refinement is that an indication with indication values having very little variation is achieved at a high rate.
In a further refinement, each pulse is given a consecutive number and the number of the pulse is stored in the position corresponding to the current time period in the circulating memory. As a result, it is no longer necessary to calculate the sum of all the numbers stored in the memory locations, but rather only the difference between the current pulse number and the oldest stored one, which is in the memory position to be written to next.
In one refinement of the invention, a first circulating memory is allocated the points in time of the pulses, and a second circulating memory is allocated the number of pulses per clock cycle. As long as the number of pulses in the second circulating memory does not reach or exceed a selected limit value, the rate is determined according to a method described above, otherwise according to a second of the methods described above. The effect achieved by this is that the averaging time is different at a relatively low rate than at a relatively high rate; for example inversely proportional to the pulse rate at a relatively low rate and constant at a relatively high rate. As a result, a relatively fast and well smoothed indication is achieved at a relatively low rate and a relative improvement in the smoothing without a disturbing delay in the indication is achieved at a relatively high rate. The limit used for changing over between the two methods may be selectable by the user of the method. For many applications, a value corresponding to twice the natural radiation will be suitable for this purpose. This value may also be adaptively incorporated by a device.
The resolution that can be attained for the rate depends on the sensitivity of the sensor and on the averaging time used. In the case of a logarithmic bar representation, therefore, the indicated bar length becomes smoother and smoother, the greater the rate is. For a numerical indication, an indication resolution corresponding to the averaging time currently used in each case is selected in one refinement of the invention.
The invention makes it possible to read at any time from the length, displayed in the indication, of a single indication bar both the current rate and the permissible residence duration, for example 2 millisieverts per hour for the rate and 1 day for the permissible residence duration. This applies even directly after the switch-on, without unreasonable variation of the indication occurring. In this case, a large range can be displayed appropriately.
If the method according to the invention is realized in a discrete signal processing technology, then it makes suitable signals available for the continual logging of pulse rates and total number of pulses with the assigned reference times.
If a vehicle moving past is regarded as the event to be recorded and an induction loop, for example, is used as the sensor, then the method can be applied particularly well in traffic counting for the purpose of assessing the road traffic density.
Further advantageous refinements emerge from the following description of exemplary embodiments. In the drawings: